Cluster Analysis on League of Legends Champions

Please follow the link below to the project:

Project 1: Cluster Analysis on League of Legends Champions

Project overview: League of Legends is a multiplayer online battle arena game developed by Riot Games where two teams fight one another to obtain a victory. Within the game there are over 150 playable characters, each grouped into 1 of 7 classes (Classes are groups of characters with similar playstyles). The overall goal of this project was to use data about each character alongside K-Means clustering to correctly classify them into their corresponding classes. In doing so the current classification system can be validated and areas of improvement identified. The results are presented in the form of tables and visualisations. This projected was completed using R.

Skills covered: Exploratory Data Analysis (EDA), Cleaning data, JSON files, Webscraping, Joining data, Data Visualisation, Data manipulation and transformation, Clustering, Principal component analysis (PCA)

Cluster Analysis on League of Legends Champions - Interactive Dashboard

Please follow the link below to the project:

Project 2: Cluster Analysis on League of Legends Champions Interactive Dashboard

Project overview: Following on from my first project, I decided to create an interactive dashboard summarising the data and the results. I believe allowing users to interact with the data themselves allows them to develop their own conclusions whilst reinforcing the findings of the project. Additionally, it improves the users engagement with the project and thus increases the likelihood that they will actively think about the conclusions and the methods as they explore the data.

Within this dashboard the user can explore my own findings, additionally, the user is allowed to create their own clusters and observe how the data is separated.

Skills covered: Interactive Data Visualisation, Data manipulation and transformation, Clustering, Rshiny, Dashboards)

Interactive Dashboard Using Rshiny

Please follow the link below to the project:

Project 2: Interactive Dashboard Using Rshiny

Project overview: This project was completed as part of an assigned university module. Within the project Rshiny is used to create an interactive dashboard for the video game Mario Kart 8. Within the game there are 32 characters who can choose from over 40 vechicles and each vechicle can be enhanced with tyre modifications. The overal aim of this dashboard was to provide an interactive environment for players of the game to explore the characters, vehicles and how the modifications influence different variables within the game.

Skills covered: Rshiny, HTML, Data aggregation

University Assignment - Data Visualisations

Please follow the link to the code and summarised version of this project:

Project 3: Summary and code

A PDF version with more detailed write up: Project 3: Detailed PDF

Project overview: This projected was completed as part of an assigned module. Within the project I take an untidy text format dataset and transform it into a format suitable for plotting. I then generated a variety of visulisations of the data and discuss and interpret the results.

Skills covered: Data visualisation, Data manipuilation and transformation

University Assignment - Regression Modelling

Please follow the link to the PDF write up:

Project 4: Regression Modelling

Project overview: Within this report linear and logistic regression techniques are applied with the goal of estimating socioeconomic determinants of a child’s nutritional status within the country of Tanzania. The report contains detailed explanations of the two methods and interpretations of the outputs of the techniques. The analysis was conducted within STATA.

Skills covered: STATA, Feature creation, Linear regression, Logistic regression, Model diagnostics

University Assignment - Data Science Foundations

Please follow the link to the PDF write up:

Project 5: Data Science Foundations

Project overview: Within this report an exploratory data analysis is carried out on a provided survey dataset. This survey measured variables related to the respondents life satisfaction. From this exploration and interesting pattern was identified between the respondents ethnicity, highest level of qualification and life satisfaction. Then linear regression is used to model one of the variables. This model is then used to predict the modeled variable.

Skills covered: Exploratory data analysis, Survey data analysis, best subset selection, model validation, k-fold cross-validation, model interpretation

Awards

I was awarded the Centre for Environmental Science Prize for best Individual Project for my undergraduate dissertation.

Within this project I used stepwise linear regression to model the drivers of deforestation within Kenya. Academic feedback from this dissertation focused on my ability to effectively and engagingly explore the story within the data.